OBSERVATIONS ON THE MATURATION PROCESSES. 29 



ate between a and e, for the stages of nuclear metamorphosis illustrated 

 in figs. 5, 6, 7, and 7a, which exhibit these forms of chromosomes (b and 

 c) are in other respects intermediate between the conditions shown in 

 fig. 4 and those of figs. 8 and 9, which present respectively the forms of 

 chromosomes shown diagrammatically in a and / (plate A, fig. G). 



Owing to the lack of exact synchronism in the formation of the 

 chromosomes, it is impossible to say with certainty which of the forms 

 b and c precedes the other, or even to assert that they are not independ- 

 ent of each other. If they do represent successive conditions of one and 

 the same chromosome, it might be imagined that the condition b had 

 been brought about by a secondary union of the four parts of such a 

 chromosome as is shown in a, followed by a bending in the equatorial 

 region, and that the condition c was afterwards reached simply by a 

 thickening of the chromosome in the region of the bending; but, on the 

 other hand, the reverse sequence might have occurred, and it may be 

 urged in support of this view that c and b represent respectively the 

 stages e and g, differing from the latter chiefly in the obliteration of the 

 cross-division, the one corresponding with the equator of the spindle. 

 As the sequence e g seems the more natural one for those two forms, 

 so in the former the sequence c b would be a natural inference. The 

 basis for the conclusion that the forms b and c pass through a stage cor- 

 responding to a is the apparent absence of those forms (6 and c) in the 

 earlier stages of nuclear metamorphosis and the prevalence of the a con- 

 dition. It must, however, be borne in mind that this does not amount 

 to a demonstration, and that individual variations in eggs or slight differ- 

 ences in preservation may afford the real explanation of the conditions. 



In b and c the transverse division of the earlier stage, a, has, then, 

 either vanished by fusion, or has not yet appeared, whereas the longi- 

 tudinal one is quite evident (plate A, fig. G, d, and plate 1, figs. 7 and 7a). 

 At the ends of the chromosome, where the spindle fibers are attached (d), 

 the chromatin is less deeply stained, as also in /. The change from the 

 condition seen in d to that of / is accomplished either by the reappearance 

 of the transverse division, or, in case it had not existed in the fundament, 

 by the first appearance of a cross-division. There is no reason, however, 

 to suppose that the form f might not in some cases arise directly from a, 

 the transverse division never being obscured. As figs. 4, 5, and 6 (plate 1) 

 show, some chromosomes develop more rapidly than others. 



The -individual chromosomes differ somewhat in size and all seem 

 to become a little smaller as they approach completion. They are at 

 first distributed over the surface of the spindle only. After they have 

 become concentrated in the region of the equatorial plane, some are 

 still found at the surface, but others are in the interior of the spindle. 

 Even at the beginning of metakinesis all do not lie exactly in the 

 equatorial plane (fig. 136). For this reason in cross-sections of spindles 

 many of the chromosomes are cut in two; polar views of the "equatorial 

 3 



